Paper Information

Title: 

LOCAL NEURAL CLASSIFIER IN MENTAL TASKS RECOGNITION

Type: PAPER
Author(s):
 
 
 
Name of Seminar: IRANIAN CONFERENCE ON BIOMEDICAL ENGINEERING
Type of Seminar:  CONFERENCE
Sponsor:  ANJOMANE MOHANDESI PEZESHKI IRAN
Date:  2004Volume 11
 
 
Abstract: 

THE MOST IMPORTANT PART OF A BCI SYSTEM IS A CLASSIFIER ASSIGNING EACH SEGMENT OF EEG SIGNAL TO ITS APPROPRIATE CLASS AMONG THE PREDEFINED CLASSES OF MENTAL TASKS, ACCORDING TO THE FEATURES EXTRACTED FROM THAT SEGMENT.
LOCAL NEURAL CLASSIFIER HAS SHOWN A GOOD PERFORMANCE IN ONLINE 3-TASK DETECTION EXPERIMENTS WITH FEEDBACK, HAVING A HIGH SPEED OF RESPONSE (EVERY 0.5 SECOND), AND A VERY LOW PERCENTAGE OF FALSE DETECTIONS (LESS THAN 5%), WHICH COMPENSATE IT’S NOT SO HIGH RATE OF CORRECT DETECTION (ABOVE 70%). HERE WE CLASSIFY RIGHT- AND LEFT-HAND MOVEMENT IMAGINATION TASKS USING LOCAL NEURAL CLASSIFIER, AND SOME OTHER TECHNIQUES SUCH AS LINEAR CLASSIFICATION, AND SSP (SIGNAL-SPACE PROJECTION), AND COMPARE THE RESULTS.
RESULTS SHOW LOCAL NEURAL CLASSIFIER IS APPROPRIATE FOR OFFLINE RECOGNITION OF THE TWO IMAGERY MOVEMENTS, CLASSIFYING GRAZ DATA. SO IT IS PROPOSED TO DESIGN AN ONLINE 3- CHANNEL BCI USING THIS METHOD, AND COMPARE THE RESULTS WITH THE RESULTS OF CLASSIFICATION ACCORDING TO PROBABILITY YK (X) CONSIDERING THE MEAN OF EACH CLASS AS THE PROTOTYPE, WHICH SHOWS BETTER RESULTS IN OFFLINE SYSTEM. AS WELL, THE BENEFITS OF FEEDBACK IN BCI ARE SHOWN BY THE RESULTS.

 
Keyword(s): BCI, EEG SIGNAL, MENTAL TASK CLASSIFICATION, LOCAL NEURAL CLASSIFIER, SSP
 
Yearly Visit 17   tarjomyar
 
Latest on Blog
Enter SID Blog